Enterovirus 71 mediates cell cycle arrest in S phase through non-structural protein 3D

Abstract

Many viruses disrupt the host cell cycle to facilitate their own growth. We assessed the mechanism and function of enterovirus 71 (EV71), a primary causative agent for recent hand, foot, and mouth disease outbreaks, in manipulating cell cycle progression. Our results suggest that EV71 infection induces S-phase arrest in diverse cell types by preventing the cell cycle transition from the S phase into the G2/M phase. Similar results were observed for an alternate picornavirus, Coxsackievirus A16. Synchronization in S phase, but not G0/G1 phase or G2/M phase, promotes viral replication. Consistent with its ability to arrest cells in S phase, the expression of cyclin A2, CDK 2, cyclin E1, and cyclin B1 was regulated by EV71 through increasing transcription of cyclin E1, promoting proteasome-mediated degradation of cyclin A2 and regulating the phosphorylation of CDK 2. Finally, a non-structural protein of EV71, the RNA-dependent RNA polymerase 3D, was demonstrated to mediate S-phase cell cycle arrest. These findings suggest that EV71 induces S-phase cell cycle arrest in infected cells via non-structural protein 3D, which may provide favorable conditions for virus production.

Introduction

Hand, foot, and mouth disease (HFMD) is generally a febrile exanthematous disease prevalent in children younger than 5 y of age. The main symptoms of HFMD are vesicles on the palmar and plantar surfaces of the hands and feet, buccal mucosa, tongue, and buttocks. Severe clinical symptoms include acute flaccid paralysis, pulmonary edema, myocarditis, and encephalitis, sometimes accompanied by death in young children.^1,2 Since first being described in California in 1969,³ many large outbreaks of HFMD have been reported in Asia, including recent outbreaks in the Fuyang,⁴ Henan,⁵ Nanchang,⁶ Hubei⁷ and Jilin¹ provinces of China, as well as Taiwan, Japan, and Malaysia.^8,9 Unfortunately, vaccines and drugs for HFMD prevention and therapy are not available, largely because the pathogenic mechanism of HFMD has not been elucidated.

Human enterovirus 71 (EV71) is a primary causative agent for HFMD that is associated with the recent outbreaks in Asia.^1,2,6 EV71 belongs to the Enterovirus genus of the Picornaviridae family, which has a single-stranded, positive-sense RNA genome of about 7400 bp consisting of 5′ and 3′ non-translated regions flanking a large open reading frame that encodes a polyprotein of about 2193 amino acids. In host cells, this polyprotein is further cleaved into 4 structural (VP1 to VP4) and 7 non-structural (2A to 3D) proteins via the virus-encoded non-structural 2A and 3C proteases.¹⁰ In addition to the 2A and 3C proteins, non-structural 3D protein is an RNA-dependent RNA polymerase that plays an important role in virus replication via the incorporation of nucleotides during RNA elongation.¹¹ Recent studies have demonstrated that 2A,¹², 3C,¹³ and 3D¹⁴ exert other roles that affect the life cycle of the virus.

As part of their pathogenic mechanism, many viruses facilitate their own growth by interacting with genes that regulate the host cell cycle. Examples can be found among DNA viruses, retroviruses, and RNA viruses. Although the DNA viruses, which replicate in the nucleus of host cells, have been the most extensively studied with regard to cell cycle control, some small DNA viruses such as simian virus 40,¹⁵ adenovirus ^16,17 and human papillomavirus,¹⁸ which lack their own polymerases, encode proteins that promote the entrance of cells into the S phase from the G1 phase by using host polymerase. Other large DNA viruses, such as herpesviruses, can induce cell cycle arrest in the G0/G1 phase to avoid competition for cellular DNA replication resources.¹⁹ As is true for DNA viruses, cell cycle regulation has been observed for retroviruses, which also replicate in the nucleus. The Vpr protein of human immunodeficiency virus type1 is responsible for eliciting cell cycle arrest in the G2/M phase.^20,21 Furthermore, RNA viruses, whose primary site of replication is normally the cytoplasm, have also been demonstrated to interfere with the host cell cycle. In the coronavirus family, infectious bronchitis virus (IBV) induces an S and G2/M-phase arrest to favor viral replication^22,23; and mouse hepatitis virus (MHV)²⁴ and some severe acute respiratory syndrome coronavirus (SARS-CoV) proteins can induce cell cycle arrest in the G0/G1 phase.^25,26

Normally, the cell cycle is controlled by the binding of CDK to the corresponding cyclin regulatory subunits, which are tightly regulated temporally. The G1 phase cyclins, cyclin D and cyclin E, associate predominantly with CDK 4/CDK 6 and CDK 2, respectively, to promote G1 progression and S-phase entry.²⁷ Both cyclin A and cyclin E then combine, mainly with CDK 2, to promote S-phase progression.²⁸ Subsequently, CDK 1 and cyclin B-forming maturation-promoting factor (MPF) regulate mitotic entry.^29,30 Some viruses have been reported to regulate cell cycle progression by manipulating cyclin and CDK progression,^31,32 but the potential effect of EV71 is unknown.

In the present study, we examined the potential effects of EV71 infection on the cell cycle. Our data show that EV71 replication induces cell cycle arrest in S phase, and, conversely, that cells arrested in S phase produce favorable conditions for the production of EV71. We further demonstrated that the non-structural 3D protein is responsible for the S-phase arrest. These results further our understanding of the pathogenic mechanisms of EV71 and provide a potential target for the treatment and prevention of HMDF disease.

Results

EV71-infected cells accumulate in S phase

To address whether EV71 affects the cell cycle of host cells, human rhabdomyosarcoma RD cells were infected with EV71 at an MOI of 1. The cells were collected at 30 h post-infection, and the cell cycle distribution was analyzed by flow cytometry. An obvious accumulation in the S phase was observed by ModFit analysis, with an increase from 37.37 ± 1.35% to 45.38 ± 0.15% for EV71-infected as compared to mock-infected cells (22.20% increase; Fig. 1A). The S-phase arrest after EV71 infection was most apparent at 24 h to 30 h post-infection (Fig. 1B). These data suggest that EV71 infection induces S-phase accumulation.

Figure 1.

S-phase accumulation induced by enterovirus 71 (EV71) infection. (A) Top panel: RD cells were mock-infected (mock) or infected with EV71 (infected) at an MOI of 1. At 30 h post-infection, cells were collected, and the cell cycle profiles were analyzed by flow cytometry. Bottom panel: The histograms were analyzed by the ModFit LT program to determine the percentage of cells in each phase of the cell cycle. ***P < 0.001. (B) The percentage of cells in each phase of the cell cycle was determined at different time points for the mock-infected or infected cells, as in panel A. The results indicate the mean ± S.D of one experiment and are representative of 3 independent experiments.

Accumulation in S phase of EV71-infected cells occurs in a variety of host cell types

To determine whether or not S-phase arrest is exclusive to the RD cell line, we selected a human glioblastoma cell line, A172, and an African green monkey kidney cell line, Vero, for further analysis of cell cycle progression after EV71 infection. A172 cells in S phase were increased from 18.10 ± 0.30% to 28.25 ± 0.82% (56.08% increase: Fig. S1A), and Vero cells were increased from 34.89 ± 1.03% to 52.68 ± 1.88% (50.96% increase; Fig. S1B) compared to EV71 infection. Thus, although A172 and Vero cell lines originated from different species and have different tissue origins, EV71 can induce S-phase accumulation in both cell lines. These results suggest that the effects of EV71 in inducing S-phase arrest are broadly applicable.

Coxsackievirus A16 (CA16) induces S-phase arrest

The CA16 virus is an additional major pathogen associated with HFMD.⁵ To determine whether cell cycle arrest also is promoted by CA16, RD cells were infected with CA16 for 30 h. CA16 virus had an effect similar to that of EV71, promoting an increase in S-phase accumulation, from 39.5 ± 0.97% to 46.59 ± 2.59%. Furthermore, after 16 h of nocodazole treatment, 24.98 ± 0.33% of the CA16-infected cells remained in S phase, as compared to 17.45 ± 0.42% of the mock-infected cells (Fig. S2A and B). These data indicate that S-phase arrest of the host cell cycle may be a common strategy for HFMD viruses.

EV71 infection prevents entry into the G2/M phase

To further understand the effects of EV71 on the cell cycle, RD cells were synchronized in G0/G1 by serum starvation for 48 h and then mock-infected or infected with the virus for 2 h. The cells were then stimulated with 10% FBS to trigger cell cycle re-entry. After 24 h of mitogenic stimulation with serum, the mock-infected cells progressed gradually into the normal cell cycle. In contrast, the majority of the EV71-infected RD cells remained in S phase over the 24-h time period (Fig. 2).

Figure 2.

EV71 infection prevents cell re-entry into G2 phase. (A) Top panel: RD cells were serum-starved for 48 h and then mock-infected (mock) or infected with EV71 (infected) at an MOI of 1. After 2 h of virus adsorption, medium containing 10% FBS was added to the cells. The cell cycle profiles were determined by flow cytometry at the indicated times post-infection. Bottom panel: The histograms were analyzed by the ModFit LT program to determine the percentage of cells in each phase of the cell cycle. The results indicate the mean ± S.D of one experiment and are representative of 3 independent experiments.

To study this phenomenon in greater detail, we treated RD cells at 8 h post-infection with 25 ng/ml nocodazole (an inhibitor of cell cycle progression to G2/M phase, through the disruption of mitotic spindles), and then analyzed the cells by flow cytometry after another 16 h. The ratio of G2/M phase cells was increased significantly by nocodazole treatment. Furthermore, only 15.68 ± 0.41% of the mock-infected cells remained in S phase after 16 h of nocodazole treatment, whereas 23.67 ± 1.21% of the EV71-infected cells remained in S phase (P<0.001; Fig. 3A). To verify these results, we also treated RD cells at 8 h post-infection with 2 μM pseudolaric acid B (PAB), an alternate inducer of mitotic arrest ^29,30; we found that only 22.56 ± 0.98% of the mock-infected cells remained in S phase after 16 h of PAB treatment. However, 33.60 ± 0.55% of the EV71-infected cells remained in S phase (P<0.001; Fig. 3B). Collectively, these results support a model in which EV71 infection induces S-phase arrest by preventing entry into the G2/M phase.

Figure 3.

Synchronization in the G2/M phase inhibits S-phase entry into infected cells. (A) Top panel: RD cells were mock-infected (mock) or infected with EV71 at an MOI of 1 (infected). At 8 h post-infection, cells were treated with 25 ng/ml nocodazole (noco), and cell cycle profiles were determined by flow cytometry at 16 h. Bottom panel: The histograms were analyzed by the ModFit LT program to determine the percentage of cells in each phase of the cell cycle. (B) Top panel: RD cells were mock-infected (mock) or infected with EV71 at an MOI of 1 (infected). At 8 h post-infection, cells were treated with 2 μM pseudolaric acid B (PAB), and cell cycle profiles were determined by flow cytometry at 16 h post-treatment. Bottom panel: The histograms were analyzed by the ModFit LT program to determine the percentage of cells in each phase of the cell cycle. The results indicate the mean ± S.D of one experiment and are representative of 3 independent experiments. ***P < 0.001.

S-phase arrest depends upon the replication of EV71

To examine whether cell cycle arrest by EV71 is dependent on viral replication, we irradiated EV71 with 0.36 J of UV light, which is known to allow EV71 to infect cells but to block its replication.³³ To determine the subsequent effects on the cell cycle, we mock-infected cells or infected them with intact virus or UV-treated virus for 8 h and then with nocodazole as described above. A significant S-phase arrest was detectable only in the cells infected with intact virus (Fig. 4A). As verification that both active and inactive virus could enter the cells, we conducted real-time PCR for the viral VP1 mRNA at 2 h post-infection, which demonstrated that cells infected with active virus and inactive virus expressed similar levels of Ct, lower than that of mock-infected cells (Fig. 4B). Furthermore, the cells infected with the UV-inactivated virus did not express the VP1 protein, whereas VP1 was expressed normally in the cells infected with intact virus, confirming that UV treatment blocks viral replication (Fig. 4C). These results suggest that S-phase cell cycle arrest by EV71 requires viral replication.

Figure 4.

S-phase arrest is required for EV71 replication. (B) Top panel: At 8 h post-infection, the cells were treated with 25 ng/mL of nocodazole (noco). Cell-cycle profiles were tested by flow cytometry at 16 h post-treatment. Bottom panel: The histograms were analyzed by the ModFit LT program to determine the percentage of cells in each phase of the cell cycle. Results indicate the mean ± S.D of one experiment and are representative of 3 independent experiments. ***P < 0.001. (B) To confirm that UV-inactivated virus infects host cells, cellular viral VP1 mRNA was detected by real-time PCR at 2 h post-infection. Relative Ct values given are the cellular viral VP1 Ct minus cellular GAPDH. (C) To ensure that UV-inactivated virus was unable to replicate, total viral VP1 from mock-infected, EV71-infected, and UV-inactivated virus-infected cells was examined by Western blotting. Histone is shown as a loading control. The results are representative of 3 independent experiments.

S phase synchronization, but not G0/G1 or G2/M phase synchronization, facilitates viral protein expression

The data outlined above indicate that EV71 induces S-phase cell cycle arrest. However, the data do not tell us whether this viral strategy is actually beneficial to the virus. To explore the possible benefits of S-phase arrest for viral replication, we infected cells with EV71 at a low MOI (0.1 MOI) and then synchronized the cells in S phase using 0.85 mM thymidine for 24 h (Fig. 5A). The expression of VP1 in S phase-synchronized cells was higher than in control non-synchronized cells at 26 h post-infection (Fig. 5B). Furthermore, the 50% tissue culture infective dose (TCID50) of the S phase-synchronized cell supernatant was significantly higher than that of control cells at 32 h post-infection (Fig. 5C). To examine whether viral RNA replication and/or transcription could be influenced by cell-cycle alteration, we detected EV71 RNA levels by real-time quantitative PCR at 6 h post-infection. EV71 RNA levels in S-phase synchronized cells were significantly higher than those in the control cells (Fig. 5D). These results suggest that S-phase arrest promotes EV71 viral replication.

Figure 5.

S-phase synchronization promotes viral replication. RD cells were infected with EV71 at an MOI of 0.1 for 2 h and then treated with 0.85mM thymidine for synchronization in (A), (B), and (C). (A) Top panel: Cell-cycle profiles were determined by flow cytometry after growth in control medium (Con) or thymidine (S)-containing medium for 24 h. Bottom panel: The histograms were analyzed by the ModFit LT program to determine the percentage of cells in each phase of the cell cycle. The results indicate the mean ± S.D of one experiment and are representative of 3 independent experiments. (B) The expression of VP1 was determined by Western blot analysis following growth in control medium (Con) or thymidine (S) medium at 26 h post-infection. Histone is shown as a loading control. The data are representative of 3 independent experiments. (C) At 32 h post-infection, the progeny viruses in the supernatant were titrated using Vero cells. A relative quantitative analysis of the 50% tissue culture infective dose (TCID50/ml) is shown. The results indicate the mean ± S.D of 3 independent experiments. ***P < 0.001. (D) RD cells were pretreated with 0.85 mM thymidine for 18 h, infected with EV71 virus for 2 h without thymidine, and then re-treated with 0.85 mM thymidine to maintain S phase. At 6 h post-infection, intracellular EV71 RNA levels were detected in control medium (Con) or thymidine (S) containing medium-treated RD cells by real-time quantitative PCR. The results were standardized using GAPDH RNA as a control and normalized to 1.0 in mock-infected cells. The results indicate the mean ± S.D of one experiment and are representative of 3 independent experiments. ***P < 0.001.

To determine whether this effect is specific for S-phase arrest, we assessed the effect of G0/G1 and G2/M arrest on viral replication. For G0/G1 phase synchronization, cells were cultured in serum-free medium for 48 h (Fig. S3A). The cells were infected with 0.1 MOI of EV71 for 2 h, and fresh culture medium, with or without serum, was then added. In contrast to the results for cells arrested at S phase, the expression of VP1 in G0/G1 phase-synchronized cells without serum was lower than that in control cells with serum at 24 h post-infection (Fig. S3B). Furthermore, the TCID50 of the G0/G1 phase-synchronized cell supernatant was significant lower than that of control cells at 30 h post-infection (Fig. S3C).

For G2/M phase synchronization, we infected cells with EV71 at 0.1 MOI and then treated them with 10 ng/ml nocodazole for 24 h (Fig. S4A). The VP1 expression levels in the G2/M phase-synchronized cells were lower than in the control cells at 26 h post-infection (Fig. S4B). Furthermore, the TCID50 of the G2/M phase synchronized cell supernatant was obviously lower than that of the control group at 32 h post-infection (Fig. S4C). These results verify that S-phase arrest induces EV71 replication, whereas both G0/G1 and G2/M arrest inhibit EV71 replication.

Key molecules regulating the cell cycle in EV71-infected cells

Progression through the cell cycle is mediated by CDKs complexed with corresponding cyclins. To identify the signaling pathway and key molecules that are responsible for EV71-induced cell cycle arrest, we examined the expression profiles of host S-phase proteins by Western blotting of RD cells at 0, 12, 24, and 36 h post-infection. Among the molecules investigated, we saw a significant increase in cyclin E1 in virus-infected cells as compared to mock-infected cells at 24 h and 36 h post-infection. However, a decrease in CDK 2, cyclin A2, and cyclin B1 proteins was observed in the same time period. Furthermore, CDK 2 was shifted toward the phosphorylated form at 24 h and 36 h post-infection (Fig. 6A). Collectively, these results verify that EV71 induces modulation of the expression profile of S-phase control proteins, a finding that is consistent with S-phase arrest.

Figure 6.

Western blot and real-time PCR analysis of host cell cycle regulatory proteins. RD cells were mock-infected (Con) or infected with EV71 at an MOI of 1 (ev) and then collected at the indicated times. (A) Cyclin A2, cyclin E1, CDK 2, and cyclin B1 were detected by Western blot analysis. Histone is shown as a loading control. Results were representative of 3 independent experiments. (B) At 30 h post-infection, mRNA levels of cyclin E1, cyclin A2 and CDK 2 were detected in mock infected (Con) and EV71 infected (EV71) cells by real-time quantitative PCR. The results are standardized to GAPDH and normalized to 1.0 in mock-infected cells. The results indicate the mean ± S.D of one experiment and are representative of 3 independent experiments. *P < 0.05. (C) At 30 h post-infection cyclin A2 expression was checked after the addition of 2 μM MG 132 (proteasome inhibitor). Histone is shown as a loading control. The results are representative of 3 independent experiments. (D) At 24 h and 36 h post-infection, phosphorylation of CDK 2 at T14, Y15, and T160 were detected. Histone is shown as a loading control. The results are representative of 3 independent experiments.

We also performed RT-PCR to assess the effects of EV71 infection on the corresponding mRNAs for cyclin A2, cyclin E1, and CDK 2 at 30 h post-infection. Consistent with the results of Western blotting, the cyclin E1 mRNA level was approximately 1.5-fold higher in the EV71-infected group than in the mock-infected group. However, there were no significant differences between the virus and mock-infected groups in cyclin A2 or CDK 2 mRNA, suggesting that the regulation of the corresponding proteins may occur at the post-translational level (Fig. 6B). We inhibited proteasome activity with MG132 and found that EV71 infection-related triggering of cyclin A2 depletion was inhibited by MG132 treatment (Fig. 6C). Furthermore, EV71 infection decreased the expression of phosphorylated CDK2 T14 and CDK2 Y15, which inhibited entry into S phase, and it also increased the expression of phosphorylated CDK2 T160, which promoted S-phase entry (Fig. 6D). Therefore, EV71 infection apparently alters the post-translational modification of these proteins.

The non-structural protein 3D of EV71 mediates S-phase arrest

Based on the known role of viral non-structural proteins in cell-cycle arrest,^20,26 we assessed the effect of exogenous expression of the apoptosis-associated non-structural viral proteins 2A and 3C. The Hep G2 cell line was selected for these studies because RD cells are not easily transfected. Initial studies verified that EV71 also caused S-phase arrest in Hep G2 cells (data not shown). 3C transfection induced G0/G1 arrest, with an increase in the percentage of G0/G1 cells from 57.18 ± 0.69% to 61.95 ± 0.72% and a decrease in the percentage of S-phase cells from 29.64 ± 0.86% to 24.04 ± 0.92 (Fig. 7A). Similarly, 2A transfection induced G0/G1 arrest, with an increase from 65.17 ± 0.39% to 69.27 ± 0.53% in the percentage of G0/G1 cells (Fig. 7A), ruling out a key role for either 3C or 2A in S-phase arrest. However, the percentage of cells in S phase was increased from 34.48 ± 0.81% to 42.17 ± 0.45% after transfection with 3D, an RNA-dependent RNA polymerase (22.29 ± 1.30% increase; Fig. 7A). To further confirm the effect of 3D on S-phase arrest, we transfected various doses of 3D into the HepG2 cell line for cell cycle evaluation. 3D increased the ratio of S phase cells in a dose-dependent manner (Fig. 7B). In order to identify the signaling pathway and key molecules responsible for 3D-induced cell cycle arrest, we then examined the expression profiles of host S-phase proteins by Western blotting of HepG2 cells at 36 h post-transfection. Among the molecules investigated, we saw a significantly increased expression of cyclin E1, a decrease in the expression of phosphorylated CDK2 T14, and an increase in the expression of phosphorylated CDK2 T160 in 3D-transfected cells when compared to mock-transfected cells (Fig. 7C). Collectively, these results verify that 3D induces modulation of the expression profile of S-phase control proteins, a function that is consistent with its role in S-phase arrest. These results suggest that the non-structural protein 3D may be primarily responsible for the S-phase arrest caused by EV71.

Figure 7.

The effect of non-structural protein 3D on S-phase arrest. (A) Effect of 3C, 2A, and 3D on cell cycle progression. Top panel: Cell cycle analysis in HepG2 cells was performed 36 h after transfection with 2 μg of pEGFP-3C-HA (3C), pcDNA3.1-IRES-2A (2A), or VR1012–3D-HA (3D), or the corresponding control vectors, pEGFP–C1, pcDNA3.1-IRES, and VR1012 (mock). Bottom panel: the histograms were analyzed by the ModFit LT program to determine the percentage of cells in each phase of the cell cycle. The results indicate the mean ± S.D of one experiment and are representative of 3 independent experiments. **P < 0.01, ***P < 001. (B) Cell cycle analysis in HepG2 cells was performed 36 h after transfection with 0, 0.5, 1, or 2 μg of VR1012–3D-HA. The histograms were analyzed by the ModFit LT program to determine the percentage of cells in each phase of the cell cycle. The results indicate the mean ± S.D of one experiment and are representative of 3 independent experiments. **P < 0.01, ***P < 001. (C) Cell cycle analysis in HepG2 cells was performed 36 h after transfection with different doses of VR1012–3D-HA. The histograms were analyzed by the ModFit LT program to determine the percentage of cells in each phase of the cell cycle. The results indicate the mean ± S.D of one experiment and are representative of 3 independent experiments. **P < 0.01, ***P < 001. (D) Cyclin E1, and phosphorylation of CDK 2 at T14 and T160 were detected by Western blot analysis. Histone is shown as a loading control. The results are representative of 3 independent experiments.

Discussion

EV71 is one of the main pathogens of HFMD.^1,9 Many studies have been done in vivo and in vitro to identify the pathogenic mechanisms of EV71 and elucidate its interaction with the host, such as cell apoptosis and inflammatory factor release assays^34-36 In the present study, we showed that EV71 induces cell cycle arrest in S phase, thereby facilitating viral production. Human rhabdomyosarcoma RD, glioblastoma A172, and monkey kidney Vero cells each displayed significant S-phase arrest after EV71 infection, despite having originated from different species (human and monkey) or different organs (brain, muscle, or kidney). These results suggest that the effects of EV71 in inducing S-phase arrest occur broadly.

To determine whether S-phase arrest is induced by other HFMD pathogens, we assessed the effects of CA16, an alternate member of the Picornaviridae family.^5,6,8 CA16 also caused S-phase arrest, suggesting that other HFMD pathogens may have similar mechanisms. To our knowledge, this is the first report of S-phase arrest induced by members of the Picornaviridae family, though members of the Flaviviridae family, including hepatitis C (HCV),³⁷ dengue virus 2,³⁸ and classic swine fever,³⁹ are known to induce S-phase arrest. Notably, the Picornaviridae and Flaviviridae share the common characteristic that they are single-stranded positive-sense RNA viruses; therefore, this effect may be a common feature that provides further insight into the pathogenic mechanism of single-stranded positive-sense RNA virus replication.

A number of viruses induce cell-cycle arrest to produce a favorable environment for replication,^22,24,33,38 and thus, we have proposed here that the S-phase arrest induced by EV71 might serve as a mechanism for promoting increased viral production. To further assess this possibility, we synchronized the cell cycle in G0/G1, S, and G2/M. G0/G1 arrest and G2 arrest both resulted in decreased viral replication; however, S-phase arrest promoted viral replication. This result is consistent with previous results for Coxsackieviruses.⁴⁰ This earlier Coxsackievirus study focused on the physiological effects of the host cell-cycle status on viral replication, whereas our study provides new information about the ability of the virus itself to manipulate the cell cycle to its advantage. Furthermore, the understanding that EV71 manipulates S-phase arrest to increase viral replication suggests the possibility of developing cell cycle-modifying medicines (especially those that induce G0/G1 or G2/M arrest) as a novel strategy to treat HFMD.

Cyclin/CDK complexes regulate cell cycle progression. For example, cyclin E/CDK 2 is responsible for regulating cellular S-phase entry from G1,⁴¹ cyclin A/CDK 2 regulates S-phase progression by replacing cyclin E,^42,43 and cyclin B1/CDK 1 is involved in the mitotic process.^29,32 To understand the underlying mechanism of S-phase arrest induced by EV71 infection, we examined the expression of host cyclin A, cyclin E, CDK 2, and cyclin B1. We demonstrated that cyclin E1 expression is up-regulated, while cyclin A2 and cyclin B1 are down-regulated after EV71 infection. These findings support a model in which: 1) EV71 can enter S phase because of an up-regulation of cyclin E expression; and 2) EV71 cannot enter G2/M because of a decrease in cyclin A and cyclin B1 expression. The ability of EV71 to induce S-phase arrest both by promoting S-phase entry and preventing G2/M entry is further supported by the effects on cell-cycle release from G0/G1 synchronization induced by nocodazole and PAB. The effect on cyclin E was verified to occur at the transcriptional level; however, cyclin A and CDK 2 mRNAs were not affected by EV71 infection; therefore, we looked at their post-translational modification and confirmed inhibiting the ubiquitin-proteasome system with MG132 inhibited the degradation of cyclin A2 induced by EV71. Furthermore, EV71 infection decreased the expression of pCDK2 T14, which inhibited entry into S phase, and it increased the expression of pCDK2 T160, which promoted entry into S.⁴⁴ We concluded that multiple cellular factors related to the cell cycle are regulated at the transcriptional and post-translational levels to support viral replication.

Given the well-characterized association between viral non-structural proteins and viral replication,^20,26 we speculated that viral proteins involved in EV71-induced replication might mediate the cell cycle arrest. The 2A and 3C proteases are known to play a role in the maturation of viral protein,¹⁰ but our results demonstrated that exogenous expression of these proteins induces G0/G1 cell-cycle arrest, rather than S-phase arrest. However, exogenous expression of 3D, an RNA-dependent RNA polymerase, mediates cell cycle arrest. 3D has been reported to induce the uridylation of a few proteins^11,45 in addition to its role in viral RNA synthesis. We found that 3D over-expression increased the expression of cyclin E1 and phosphorylated CDK2 T160, which promoted S-phase entry. Therefore, our study demonstrates a new function for a 3D protein of the Picornavirus family and implicates 3D as a putative target for the strategic development of new antiviral therapies.

Materials and methods

Viruses and cells

The Changchun077 strain of EV71 has been reported previously.¹ The Shzh05 strain of CA16 (GenBank Accession No. EU262658) was a gift from Professor Qi Jin at the Institute of Pathogen Biology, Beijing. The viruses were propagated in African green monkey kidney cells (Vero). Vero (No CCL^-81), human glioblastoma A172 (No CRL-1620), human hepatocyte HepG2 (No HB-8065), and human rhabdomyosarcoma RD (No CCL^-136) cells were purchased from the ATCC (Manassas, VA, USA) and were maintained in Dulbecco's modified Eagle's medium (DMEM) (Hyclone, Logan, UT, USA) supplemented with 10% fetal bovine serum (FBS) (GIBCO BRL, Grand Island, NY, USA).

Plasmid construction and transfection

Total RNA was extracted from EV71 (Changchun077) virus stock with 200 μl Trizol (Invitrogen). cDNA was generated using the High-capacity cDNA Reverse Transcription Kit (Applied Biosystems) and oligo-d(T)18 primers (Takara) according to the supplier's instructions. One-fifth of the volume of the product from the reverse transcription reaction was used as a template for PCR of the 3D gene, using the primers given in Table 1. The forward primer was engineered to contain an HA tag, and both primers contained restriction sites for cloning. The RT-PCR product was subsequently digested and directionally cloned into the corresponding restriction sites of plasmid VR1012. pcDNA3.1-IRES-2A ¹² was a gift from Dr. Shih-Yen Lo, and pEGFP-3C ¹³ was a gift from Dr. Jianwei Wang. Following the manufacturer's protocol, we transfected 2 μg of plasmid together with 6 μl of Lipofectamine 2000 (Invitrogen) into cells in a 6-well plate; in the dose-dependent experiment with 3D, we replenished the VR1012 to maintain a concentration of 2 μg plasmid/ well in a 6-well plate.

Table 1.

The primers used in this study

Primer pair	Role	Forward sequence5′-3′(restriction enzyme)	Reverse sequence5′-3′ (restriction enzyme)
3D	Plasmid construction	CTGCAGACCTAGTACCCTTACGACGTCCCAGATTA-CGCGGGAGAGATCCAGTGGGTTA (PstI)	GGATCCCTAAAATAACTCGAGCCAATTG (BamHI)
Cyclin E	Real time	TCAGGGTATCAGTGGTGCGA	CAAATCCAAGCTGTCTCTGTG
Cyclin A	Real time	GCATGTCACCGTTCCTCCTT	CAGGGCATCTTCACGCTCTAT
CDK 2	Real time	CTCCTGGGCTCGAAATATTATTCCACAG	CCGGAAGAGCTGGTCAATCTCAGA
VP 1	Real time	AGCACCCACAGGCCAGAACACAC	ATCCCGCCCTACTGAAGAAACTA
GAPDH	Real time	GCAAATTCCATGGCACCGT	TCGCCCCACTTGATTTTGG

Viral titer determination

The viral titer was determined by measuring the TCID50 in a microtitration assay using Vero cells, as described.⁴⁶ In brief, Vero cells were seeded into 96-well plates and incubated at 37°C for 24 h. Virus-containing supernatant was serially diluted 10-fold, and 100 μl was added per well in octuplicate. The cytopathic effect was observed once per day until the experimental endpoint was reached. The viral titer of the TCID50 was determined according to the Reed-Muench method,⁴⁷ based on the assumption that material with 1 × 10⁵ TCID50/ml will produce 0.7 × 10⁵ plaque forming units/ml (www.protocol-online.org/biology-forums/posts/1664.html).

Infection

With the exception of viral production after cell cycle synchronization, which was done with a multiplicity of infection (MOI) of 0.1, cells were mock-infected or infected with EV71 at an MOI of 1. After 2 h of virus adsorption, cells were washed with PBS one time, and fresh culture medium was added.

Cell cycle release

Subconfluent cultures of RD cells were synchronized in G0/G1 phase by serum deprivation for 48 h ³³. Approximately 5 × 10⁵ cells were plated in a 6-well plate and maintained in serum-free medium for 48 h. Virus at an MOI of 1 or fresh culture medium was added for 2 h, the cells were washed once with PBS, and fresh 10% DMEM was added to release the cells from G0/G1.

Synchronization of cells

To observe the effect of the cell cycle on virus growth, subconfluent cultures of RD cells were synchronized in G0/G1 phase by serum deprivation for 48 h ³³. For S-phase synchronization, cells were treated with thymidine (Sigma) ³⁸ for 24 h at a final concentration of 0.85 mM. For G2/M arrest, RD cells were treated with nocodazole (Sigma) ³³ for 24 h at a final concentration of 10 ng/ml. For sustained S and G2/M cell-cycle arrest after virus infection, fresh 0.85 mM thymidine and 10 ng/ml nocodazole were added for the indicated times.

Cell cycle analysis by flow cytometry

Nuclear DNA content was measured using propidium iodide (PI) staining and fluorescence-activated cell sorting (FACS). Adherent cells were collected by treatment with trypsin and then washed with phosphate-buffered saline (PBS). The cells were fixed in 1 ml of cold 70% ethanol overnight at 4°C and resuspended in staining buffer (50 μg/ml PtdIns [Sigma], 20 μg/ml RNase in PBS) for 2 h at 4°C. PI-stained cells were then analyzed using FACS (FACScan; BD), and at least 10,000 cells were counted for each sample. Data analysis was performed by using ModFit LT, version 2.0 (Verity Software House).

Western blot analysis

Virus-infected and mock-infected cells were collected at various times after EV71 infection and washed once with PBS. Cells were lysed directly in sodium dodecyl sulfate (SDS) sample buffer (60 mM Tris-HCl [pH 6.8], 2% SDS, 10% glycerol, 5% 2-mercaptoethanol, 0.01% bromophenol blue), followed by boiling for 10 min. Whole-cell lysates were further subjected to SDS-PAGE. Proteins were transferred to nitrocellulose membranes (Bio-Rad) and detected with corresponding primary and alkaline phosphatase-conjugated secondary antibody. The membranes were then reacted with 5-bromo-4-chloro-39-indolylphosphate (BCIP) and nitro-blue tetrazolium (NBT) substrate (Sigma). The following mouse or rabbit antibodies were used in Western blot analyses: anti-CDK 2 (Boster), anti-CDK 2 T14 (Abcam), anti-CDK 2 Y15 (Abcam), anti-CDK 2 T160 (Cell Signal), anti-cyclin A2 (Proteintech), anti-cyclin E1 (Abcam), anti-histone (GenScript), and anti-cyclin B1 (Santa Cruz). Anti-VP1 polyclonal antibody was prepared by our laboratory. Mouse and rabbit secondary antibodies were obtained from Proteintech.

Quantitative real-time RT-PCR

All work was carried out in a designated PCR-clean area. RNA was extracted from infected and uninfected cells using Trizol reagent (Gibco-BRL, Rockville, Md.) and isolated as specified by the manufacturer. The RNA was DNAse-treated (DNase I-RNase-Free, Ambion) to remove any contaminating DNA; 200 ng of total RNA was reverse-transcribed with oligo dT primers using the High Capacity cDNA RT Kit (Applied Biosystems) in a 20-μl cDNA reaction, as specified by the manufacturer. For quantitative PCR, the template cDNA was added to a 20 μl reaction with SYBR GREEN PCR Master Mix (Applied Biosystems) and 0.2 μM of primer (Table 1). The amplification was carried out using an ABI Prism 7000 for 40 cycles under the following conditions: an initial denaturation of 95°C for 10 min, plus 40 cycles of 95°C for 15 s, then 60°C for 1 min. The -fold changes were calculated relative to GAPDH using the △△Ct method for cylin A2, cyclin E1, and CDK 2 mRNA analysis. For the inactivation virus entry test, relative Ct values were calculated by determining the VP1 Ct value minus the GAPDH Ct value. Primers are listed in Table 1.

Statistical analyses

Statistics were analyzed using Student's t-test. Data are presented as means and standard deviations (SD). *P values of <0.05, **P values of <0.01, and ***P values of <0.001 were considered statistically significant.

Disclosure of Potential Conflicts of Interest

There was no conflict of interest for any author.

Funding

This work was mainly supported by funding from National Natural Science Foundation of China (81301416). It was also supported by funding from the Chinese Ministry of Science and Technology (2012CB911100 and 2013ZX10001005), the Jilin Provincial Science and Technology Department (20140204004YY), the State Grade III Laboratory of Traditional Chinese Medicine, Immunology and Molecular Biology Laboratory of Chinese Ministry of Education (IRT1016), The Key Laboratory of Molecular Virology of Jilin Province (20102209), and a grant (2R56AI62644–6) from the NIAID.

Supplemental Material

Supplemental data for this article can be accessed on the publisher's website.